Currently, my primary research focus is in the early detection of prostate cancer using gene-specific promoter region hypermethylation. Gene hypermethylation promises to be a good early detection marker since it is a DNA-based, stable covalent modification of the DNA. My laboratory has developed assays sensitive enough to detect gene hypermethylation down to around 20 cells in a mixed pool of methylated and unmethylated DNA (can detect in a 1:10,000 ratio of methylated:unmethylated). We are able to detect tumor-specific gene methylation in serum and diagnostic biopsy samples of men with prostate cancer. We are currently initiating studies to evaluate whether evaluating methylation of a panel of genes (we are currently investigating six genes) can aid in the diagnosis of prostate cancer. The current standard of care for the diagnosis of prostate includes serum prostate specific antigen (PSA) test and ultra-sound guided biopsy. One problem associated with this regime is the relatively low sensitivity and specificity of serum PSA for cancer detection. Only about 25% of men diagnosed with cancer have elevated PSA and about 50% with elevated PSA are diagnosed with cancer. Molecular markers, in addition to PSA, that may improve the sensitivity and specificity are required. In addition, the diagnostic biopsy procedure has a 20% false negative rate. That is, 20% of men are not diagnosed on their initial biopsy and if they have persistently elevated PSA levels need to have repeat biopsy. Detection of tumor-specific gene methylation in diagnostic biopsy cores may aid in the diagnosis of prostate cancer by reducing the false negative biopsy rate. Prior studies reported the ability to detect gene methylation in urine and serum specimen of prostate cancer patients, providing a rationale for its potential use as a biomarker. Our study, the Prostate Molecular Marker Study (PMMS), was designed to address the question of the utility of gene-specific methylation for both the improvement of PSA screening and the reduction of the false negative biopsy rate. The PMMS was conducted in collaboration with Urologists at the Madigan Army Medical Center in Tacoma, WA. One reason for choosing Madigan was that being an active military site it has a returning patient population and infrastructure to follow the patients over time with no extra cost. The study enrolled 100 men that were referred for diagnostic biopsy (complete enrollment ended July 2005). At the biopsy procedure, serum and a post-prostatic massage urine specimen were collected and sent to my laboratory for methylation analysis. Tissue sections from each biopsy core (8 in total) and radical prostatectomy specimen when available were also sent for evaluation. We are in the process of finishing the DNA extractions and methylation assays of both urine and biopsy material. Another aim of my laboratory is to evaluate whether gene hypermethylation status can predict more aggressive disease. Gene methylation may predict men who may have a recurrence and these men can be treated with adjuvant therapy (they currently undergo "watchful waiting"). My laboratory is currently working on characterizing the methylation status of genes associated with advance stage and invasive disease. About 30% of men with clinically localized prostate cancer who receive a radical prostatectomy undergo biochemical recurrence (a rise in serum PSA) within 2 to 5 years after surgery. This is clinically significant since biochemical recurrence has been associated with clinical recurrence and the presence of metastatic disease. We investigated whether gene methylation has prognostic significance in a biochemical recurrence study. We were primarily interested in genes that my laboratory had identified as being associated with tumor grade (CD44 and PTGS2) since tumor grade is currently one of the strongest prognostic markers in clinical practice. We found that PTGS2 and CD44 were associated with a 9-fold increased risk for biochemical recurrence. Kaplan-Meier analysis showed men with methylation of these genes in their tumor had a statistically shorter time to recurrence. This work has been submitted for publication. I am currently exploring collaborations to validate these potentially important findings in a second (independent) recurrence study. Several institutions with ongoing recurrence studies may have DNA available for future biomarker studies